Task Scoping for Efficient Planning in Open Worlds (Student Abstract)

Authors

  • Nishanth Kumar Brown University
  • Michael Fishman Brown University
  • Natasha Danas Brown University
  • Stefanie Tellex Brown University
  • Michael Littman Brown University
  • George Konidaris Brown University

DOI:

https://doi.org/10.1609/aaai.v34i10.7195

Abstract

We propose an abstraction method for open-world environments expressed as Factored Markov Decision Processes (FMDPs) with very large state and action spaces. Our method prunes state and action variables that are irrelevant to the optimal value function on the state subspace the agent would visit when following any optimal policy from the initial state. This method thus enables tractable fast planning within large open-world FMDPs.

Downloads

Published

2020-04-03

How to Cite

Kumar, N., Fishman, M., Danas, N., Tellex, S., Littman, M., & Konidaris, G. (2020). Task Scoping for Efficient Planning in Open Worlds (Student Abstract). Proceedings of the AAAI Conference on Artificial Intelligence, 34(10), 13845-13846. https://doi.org/10.1609/aaai.v34i10.7195

Issue

Section

Student Abstract Track